diff --git a/doc/fluid/user_guides/howto/dygraph/DyGraph.md b/doc/fluid/user_guides/howto/dygraph/DyGraph.md index f3bc247c1cb7e2da40e26b4a6a15751376105d97..31395e0fa2e69aa2fc649bf7549df0373503577c 100644 --- a/doc/fluid/user_guides/howto/dygraph/DyGraph.md +++ b/doc/fluid/user_guides/howto/dygraph/DyGraph.md @@ -436,7 +436,7 @@ Dygraph将非常适合和Numpy一起使用,使用`fluid.dygraph.to_variable(x) if batch_id % 100 == 0 and batch_id is not 0: print("epoch: {}, batch_id: {}, loss is: {}".format(epoch, batch_id, avg_loss.numpy())) -修改的地方主要有三处: +动态图单卡训练转多卡训练需要修改的地方主要有四处: 1. 需要从环境变量获取设备的ID,即: place = fluid.CUDAPlace(fluid.dygraph.parallel.Env().dev_id) @@ -481,7 +481,7 @@ Paddle动态图多进程多卡模型训练启动时需要指定使用的GPU, trainers_endpoints: 127.0.0.1:6170,127.0.0.1:6171,127.0.0.1:6172,127.0.0.1:6173 , node_id: 0 , current_node_ip: 127.0.0.1 , num_nodes: 1 , node_ips: ['127.0.0.1'] , nranks: 4 -此时,程序会将每个进程的输出log导入到./mylog路径下: +此时,程序会将每个进程的输出log导出到./mylog路径下: . ├── mylog